Live data from Hacker News

Redis is open source again

antirez.com

621–630 of 815 posts

Re: Redis is open source again

#621

Earlier quoted context omitted.

"walking off with your money" This is the heart of all of these stupid takes. There is no "your money" for anyone else to walk off with. Nothing was stolen. If you want to sell software, or rent access to software, then just do that honestly forom the outset. And good luck to you on that. I will not consume it unless I have no other choice, and I will not contribute to it at all period, even tertially by for instance…

You’re living in theoretical fantasy land. Yeah, it’d be great if everything was open source and everyone was happy. But that’s not really what happens. Redis and Elastic are popular exactly because there’s entire companies behind it constantly maintaining it, securing it, and adding features. If you want use open source software at your job that’s a hard requirement. No, you can’t use “MySillyKeyValueCache” that Tim…

> Companies cost money to run. Google and AWS exploit open source work for profit, and don’t contribute back enough to ensure its support.

I don’t understand this take, if Google/AWS/Microsoft decide to host your service and some vulnerabilities are discovered then you got free Security Research done, your product would either have remained vulnerable or someone else would have filed the same bug.

You own the code so you can decide only certain people can commit PRs and can choose to close any feature requests issues.

If one of the cloud providers decided to use your product it’s because they deemed it reasonable as is or they can fork / contribute upstream.

That is what OSS is, by definition.

Hell you can post the code with a license like MIT and then never touch it again, and if someone else can monitise that code kudos to them.

If you are an existing company and open sourced your code, Facebook/react, then you presumably already make enough money to support development yourself or intend to stop development.

If you open source code that is your core business and somebody “steals your lunch” your learnt an important lesson and hopefully won’t make the same mistake twice. If you then decide to relicense and the community abandons you and causes uproar on the internet you are reaping the rewards of your actions, accept them.

Re: Redis is open source again

#622
post #575
post #521

Earlier quoted context omitted.

> Whats the point of open source if you can't fork or sell hosted clusters? The problem here is that Amazon is in a position where they can fork and sell it for a loss in order to crush the competition. That would be the point of antitrust, I guess? I mean if regulations were actually enforced.

Do AWS really sell things at a loss?

Not really. I have done multiple AWS pricing and costing exercises while launching services and there was never a direction from the VP to sell it lower than what it costs to build and operate. Cost to build and operate includes everything from salaries, infrastructure and many other line items. And usually things are projected 3-5 years into the future and P&L analysis must show that eventually the service will make a profit. It does make some assumptions about minimum customer adoption for the profit margins to materialize which eventually becomes part of the product and sales teams goals.

The costing model does allow losses to be incurred in the initial years because building the thing is more expensive at first but then it should settle down and revenue should outpace expenses.

What can happen with these open source products being launched as a service is that that initial cost can be cut down by as much as 50-75% but rarely more than that because you still need to build all the surrounding infrastructure, documentation, UI. It still gives AWS an advantage by relying on an existing body of work they can start with where many problems have been thought through and solved. Also you will likely get a good product roadmap skeleton ready to be prioritized which otherwise can be a huge time sink.

In a nutshell, no. AWS won’t sell a service at a loss (there are exceptions of course) but there is room to incur a loss at the beginning but it is priced to eventually turn a profit. Whether that happens for every service in reality is a different story.

Re: Redis is open source again

#623
post #437

I would love to see Amazon offer a Redis service, comply with the AGPLv3 instead of using the other Redis licensing options and do revenue sharing to incentivise more commercial source-available projects to go that route.

Microsoft does this.

Re: Redis is open source again

#624
post #599
post #353

Earlier quoted context omitted.

> The freedom to run the program as you wish, for any purpose I can do this with AGPL software just fine. Which purposes are not allowed under AGPL?

You are not allowed to modify an AGPL program without adding a mandatory feature to provide the currently-running source code to a remote network user of the software. You are also not allowed to remove that functionality. It's a clear violation of freedom 0. I suggest that you read the actual text of the AGPL. It's madness. They wanted a EULA so badly.

> You are not allowed to modify an AGPL program without adding a mandatory feature to provide the currently-running source code to a remote network user of the software

That's not true. You only need to provide the modified source code somewhere.

You might want to read it again :-) It seems like you have a wrong idea of what the AGPL does.

edit: I see an immediate downvote. If whoever did this thinks I'm wrong, please share the exact AGPL excerpt that mandates actually building a feature to provide the source code currently in use in the network program itself.

Re: Redis is open source again

#625
post #624
post #599

Earlier quoted context omitted.

You are not allowed to modify an AGPL program without adding a mandatory feature to provide the currently-running source code to a remote network user of the software. You are also not allowed to remove that functionality. It's a clear violation of freedom 0. I suggest that you read the actual text of the AGPL. It's madness. They wanted a EULA so badly.

> You are not allowed to modify an AGPL program without adding a mandatory feature to provide the currently-running source code to a remote network user of the software That's not true. You only need to provide the modified source code somewhere. You might want to read it again :-) It seems like you have a wrong idea of what the AGPL does. edit: I see an immediate downvote. If whoever did this thinks I'm wrong, pleas…

The problematic clause is this one:

> if you modify the Program, your modified version must prominently offer all users interacting with it remotely through a computer network (if your version supports such interaction) an opportunity to receive the Corresponding Source

The application does not have to directly provide its own exact source code, but it must advertise where to find it.

And it's quite unclear what exactly counts here. For a web application it's somewhat obvious how one might do that (although how many AGPL applications are prepared for it as a simple config option vs require you to do additional patching to add it), but for other applications its not.

E.g. you interact with redis over a custom protocol thats self-described as being human-readable. Does that mean I need to inject the offer there somewhere? What if it is a much more opaque protocol? If you ask 5 lawyers, how many opinions on that do you get?

Re: Redis is open source again

#626
post #573

Earlier quoted context omitted.

> Amazon's product had confusing naming and positioning, and lagged in compatibility which created a headache for Elastic. Irrelevant. Elastic decided to release software to the public with a license that granted everyone under the sun the right to use it how they saw fit. This included also selling services. That became a big part of their business model as it drove up it's adoption rates and popularity. If Elastic…

Yes, anti trust rules should have stopped amazon. But they didn't, this directly hurts open source. I don't think elastic would've tried the license change because of competition using the open source. The problem is Amazon, not elastic that's trying to survive.

> Yes, anti trust rules should have stopped amazon

Which ones, specifically?

Re: Redis is open source again

#627

I think a project has to be Linux or Postgres big to really operate as open source. Why? Because cloud providers can easily sell your product and contribute nothing back. So, for anything Personally, I think it is better just to call a spade a spade and release things with commercial licenses. Let all companies with $1T category. The above is a bit of a strawman but what we really want is for valuable projects to con…

What about looking at it the other way around, neither Linux nor PostgreSQL is a commercial product, never were and never will be. Many people create commercial products on top of them, and that's OK, it's allowed by the license. Even Redis started as a fully open source project. But then it was sold to a company, and company cares more about making money than taking care of open source product. If you are a small company, and want to create an open source product, that's only free for personal use, just say so up front. Don't try to pretend the product is something else. It's really not about size, it's about motivation, and the motivation of a company is always making money in the first place.

Re: Redis is open source again

#628
post #619

Earlier quoted context omitted.

User freedom means not giving someone else the freedom to take away user freedom though, which is what doormat licenses (can we call them cuck licenses or is that too far for HN?) do.

> User freedom means not giving someone else the freedom to take away user freedom though This sentence does a shortcut that's worth expanding on, because it can give the impression that permissive licenses don't respect user rights, and because it doesn't cleanly apply to every situations. User freedom is full with a permissive license. If I get software under MIT, my freedom is fully respected. But it is true that…

That is true. I'm talking general purpose. If you're putting something you made on the internet, and there's a good chance it's useful to someone, I recommend AGPL unless you have a specific argument against it. Also notice that it's easier to grant more permission than to take it away.

Re: Redis is open source again

#629
post #33

I contributed a minor (but imho still neat :p) improvement to Redis under its original license, and personally moved to using redict when the unexpected license change to SSPL was announced - and I was feeling betrayed as a contributor to a properly-FOSS-codebase. (Had they switched to AGPL right away, I'd have been perfectly fine with that change from a moral perspective, ftr.) I have a great deal of respect for ant…

Yeah, we just did this whole ride with Elastic [0]: company changes the license out from under the community, community revolts, company gives up and changes it back. Both companies even pulled the same "it worked" excuse ("while it was painful, it worked", "this achieved our goal"). Neither company has built in a legal safety mechanism to prevent themselves from pulling the rug again later and both companies have sh…

The legal safety mechanism is the license. They gave you software under a certain license and you're in the clear as long as you follow it. You don't have to delete it if they give different software under a different license.

If you do need constant updates, you may trust forks more not to switch license, but forks tend to disappear at about the same rate that originals switch license, so why does it matter? Such is the nature of relying on free stuff - you're at the mercy of the one who hands it out.

Re: Redis is open source again

#630

Earlier quoted context omitted.

I said exactly that's a win of the community. But money are needed to pay the folks that work at open source software, and the companies that went for the SSPL were trying to protect their business (and, as a side effect, wanted or not, the ability to pay for such work). I believe the software world failed to protect open source software in the cloud era, but in general the environment that we collectively created ma…

I don't buy that justification. These two companies (and the others like them) set up a model for funding their business knowing full well that they would have to compete with others who were able to provide the same service they were. That's always been baked into FOSS—you can't plan around a monopoly when you're releasing your code for free, that's part of the trade-off you make. When AWS and Co did provide hosting…

> now it's suddenly AWS who are the bad guys for using the FOSS software the way the license always said they could

That's not a fair representation of the objection. The thing under contention here is a behemoth freeloading for substantial profit. The reason it's viewed as bad is because that behavior undermines the community at large. Participants need to contribute back proportional to their benefit when realistically able lest the community decline over time.

It's conceptually similar to the disapproval employers with poor workplaces are subjected to. Even if they technically obey all the labor laws, by prioritizing their profits over worker health and well being they garner a bad reputation. "But we didn't do anything illegal" isn't going to get them out of it.

Post reply on HN